Benchmarking Hierarchical Script Knowledge

@inproceedings{Bisk2019BenchmarkingHS,
  title={Benchmarking Hierarchical Script Knowledge},
  author={Yonatan Bisk and Jan Buys and Karl Pichotta and Yejin Choi},
  booktitle={NAACL-HLT},
  year={2019}
}
  • Yonatan Bisk, Jan Buys, +1 author Yejin Choi
  • Published in NAACL-HLT 2019
  • Computer Science
  • Understanding procedural language requires reasoning about both hierarchical and temporal relations between events. For example, “boiling pasta” is a sub-event of “making a pasta dish”, typically happens before “draining pasta,” and requires the use of omitted tools (e.g. a strainer, sink...). While people are able to choose when and how to use abstract versus concrete instructions, the NLP community lacks corpora and tasks for evaluating if our models can do the same. In this paper, we… CONTINUE READING
    5 Citations

    Figures, Tables, and Topics from this paper.

    PIQA: Reasoning about Physical Commonsense in Natural Language
    • 36
    • PDF
    Reading Between the Lines: Exploring Infilling in Visual Narratives
    Event Representation with Sequential, Semi-Supervised Discrete Variables
    • 1
    • PDF
    Integrating Text and Image: Determining Multimodal Document Intent in Instagram Posts
    • 12
    • PDF

    References

    SHOWING 1-10 OF 38 REFERENCES
    Unsupervised Learning of Narrative Event Chains
    • 442
    • PDF
    Generating Coherent Event Schemas at Scale
    • 78
    • PDF
    Probabilistic Frame Induction
    • 88
    • PDF
    Event Schema Induction with a Probabilistic Entity-Driven Model
    • 80
    • PDF
    Behind the Scenes of an Evolving Event Cloze Test
    • 6
    • PDF
    Deep contextualized word representations
    • 4,601
    • PDF